Update README.md

README.md

@@ -18,6 +18,19 @@ March 2023
 * Blog post
 * Github (mosaicml/examples repo)
 
+# How to use
+
+```python
+from transformers import AutoModelforForMaskedLM
+mlm = AutoModelForMaskedLM.from_pretrained('mosaicml/mosaic-bert-base', use_auth_token=<your token>, trust_remote_code=True)
+```
+The tokenizer for this model is the Hugging Face `bert-base-uncased` tokenizer.
+
+```python
+from transformers import BertTokenizer
+tokenizer = BertTokenizer.from_pretrained('bert-base-uncased')
+```
+
 ## Model description
 
 In order to build MosaicBERT, we adopted architectural choices from the recent transformer literature. 
@@ -60,22 +73,8 @@ reduces the number of read/write operations between the GPU HBM (high bandwidth
    MosaicBERT-Base trains faster than BERT-Base despite having more parameters. 
 
 
-# How to use
 
 
-
-
-```python
-from transformers import AutoModelforForMaskedLM
-mlm = AutoModelForMaskedLM.from_pretrained('mosaicml/mosaic-bert-base', use_auth_token=<your token>, trust_remote_code=True)
-```
-The tokenizer for this model is the Hugging Face `bert-base-uncased` tokenizer.
-
-```python
-from transformers import BertTokenizer
-tokenizer = BertTokenizer.from_pretrained('bert-base-uncased')
-```
-
 ## Training data
 
 MosaicBERT is pretrained using a standard Masked Language Modeling (MLM) objective: the model is given a sequence of